What percentage of Russian seaborne crude oil export loading capacity will be offline at end of June 2026 due to Ukrainian strikes (vs the H2 2025 baseline)?

TL;DR

My resolution forecast is 13% offline, with the modal resolution near 11.6%. June flow data shows the late-March 40% outage did not persist: Russia was loading record or near-record crude volumes at sea. The main source-path risk is Bloomberg's exact June 1–28 tanker number, which maps to 17.4%, versus the CREA/Kpler-style fallback near 11.6%.

Context

The resolution window has already closed, but the primary-source deadline has not: the question waits until 14 July 2026 for a Reuters or Bloomberg stocktake that explicitly quantifies crude export loading capacity offline due to Ukrainian operations. I found flow articles, not a clean end-June capacity-offline stocktake. That makes the fallback path the dominant resolution path.

The late-March shock was real. Reuters reported that about 2.0 mb/d, or 40%, of Russian crude export capacity was offline after strikes and related disruptions at Primorsk, Ust-Luga, Novorossiysk, Druzhba, and tanker flows (Reuters via The Moscow Times, 25 Mar. 2026; Reuters via The Business Standard, 2 Apr. 2026). By June, the story had changed: Bloomberg said Baltic and Black Sea export terminals had largely resumed normal operations by the four weeks to 31 May, and Reuters said western-port June crude shipments from Primorsk, Ust-Luga, and Novorossiysk reached nearly 3.0 mb/d (Bloomberg via The Moscow Times, 2 Jun. 2026; Reuters via Stockopedia, 2 Jul. 2026).

Evidence

The baseline supports the 5.0 mb/d denominator but not a 5.0 mb/d normal-flow assumption. KSE's April assessment, using Kpler data accessed on 7 April 2026, put 2025 observed oil export flows at 1.26 mb/d through Primorsk, 1.15 mb/d through Ust-Luga, 0.97 mb/d through Novorossiysk, 1.00 mb/d through Nakhodka/Kozmino, and 0.30 mb/d through Murmansk, or 4.68 mb/d across the main nodes (KSE Institute, 9 Apr. 2026). The same report shows why March mattered: Ust-Luga crude fell from 725 kb/d in the week of 16–22 March to 101 kb/d and 155 kb/d in the next two weeks, while Primorsk crude stayed near 780 kb/d and then rose to 848 kb/d (KSE Institute, 9 Apr. 2026). Bloomberg's tanker tracking also showed a one-week drop from 4.07 mb/d to 2.32 mb/d in the week to 29 March, driven by Baltic disruptions (Bloomberg via The Moscow Times, 31 Mar. 2026).

The comparable flow history below uses average million barrels per day of Russian seaborne crude, not products. Annual rows are calendar-year averages, monthly rows are calendar months, and four-week rows are 28-day tanker-tracking averages; the source vintage is the publication date shown in the source column.

The fallback formula is:

$$R = \max\left(0, \left(1 - \frac{E}{5.0}\right) \times 100\right),$$

where $E$ is average Russian seaborne crude exports in mb/d. CREA's June bulletin says Russian crude export volumes rose 14% month on month, with Novorossiysk crude loadings up 68%, Ust-Luga down 10%, and Primorsk down 9% (CREA, 10 Jul. 2026). Applying that 14% rise to KSE's May seaborne crude figure of 3.88 mb/d gives 4.42 mb/d, which implies 11.5% under the fallback; S&P Global's separate 4.414 mb/d June estimate implies 11.7% (KSE Russian Oil Tracker, 1 Jul. 2026; S&P Global CAS, 2 Jul. 2026, updated 3 Jul.). Bloomberg's exact-window 4.13 mb/d implies 17.4%, and I keep substantial probability there because it matches the 1–28 June window exactly (Bloomberg via Meduza, 30 Jun./1 Jul. 2026).

The June narrative flipped during the month. Reuters reported on 8 June that western-port crude loadings could fall to 1.7 mb/d in June from 2.5 mb/d in May as Russia tried to raise refinery runs and output weakened (Reuters via MarketScreener, 8 Jun. 2026). By 24 June, Reuters sources expected those same ports to reach 2.7–2.8 mb/d because refinery damage left more crude available for export, and by 2 July Reuters sources put western-port exports at nearly 3.0 mb/d in June (Reuters via Ukrainska Pravda, 24 Jun. 2026; Reuters via Stockopedia, 2 Jul. 2026). CREA also said seaborne oil-product loadings fell 21% month on month to a record low, which supports the same mechanism: refineries and product logistics were the binding damage, not crude loading berths (CREA, 10 Jul. 2026).

My distribution puts 55% on the CREA/Kpler/S&P-style fallback near 11.6%, 22% on the Bloomberg exact-window path near 17.4%, 8% on a source blend near 14.6%, 10% on a late qualifying physical-capacity stocktake near 6–8%, and 5% on higher source or attribution surprises. That gives a mean near 13%, a median near 12%, and an 80% interval of roughly 10–18%. The physical engineering outage in June was probably lower than the resolved fallback value, because western ports were loading close to capacity while the fallback compares realized exports with a 5.0 mb/d capacity denominator.

What's non-obvious

The obvious story from March was that Ukraine had found a durable way to choke Russian crude exports. June data says the opposite for crude. Ukraine was still damaging the oil system, but refinery strikes pushed more unrefined oil to export terminals; Bloomberg described higher seaborne crude exports as a result of lower refinery processing, and Reuters reported that western-port loadings were revised up for the same reason (Bloomberg via The Moscow Times, 2 Jun. 2026; Reuters via Ukrainska Pravda, 24 Jun. 2026).

The resolution rule is also a trap. A normal 2025 seaborne crude export month of 3.47 mb/d would mechanically print 30.6% under the fallback even with no port damage, because the formula compares flow to a 5.0 mb/d capacity baseline (IEA/Kpler/Argus May OMR mirror, 13 May 2026). June was not normal; it was a record or near-record crude export month, so the fallback falls into the low-to-mid teens instead of the 30% range (Bloomberg via Meduza, 30 Jun./1 Jul. 2026; S&P Global CAS, 2 Jul. 2026, updated 3 Jul.).

Limitations

The largest uncertainty is source-path risk. I did not find a Reuters or Bloomberg article, as of 12 July 2026, that explicitly quantifies June-average crude export loading capacity offline due to Ukrainian operations in the same way the 25 March Reuters article did. If such a qualifying article appears or is found by 14 July 2026, it would supersede the fallback and could pull the answer toward a lower physical-capacity estimate.

The second gap is CREA's public text. CREA gives the 14% month-on-month crude-volume rise and port-level changes, but the accessible article text does not show a single all-Russia seaborne crude mb/d number (CREA, 10 Jul. 2026). I infer the CREA/Kpler-style value from KSE's May 3.88 mb/d and cross-check it against S&P Global's 4.414 mb/d June print, but Bloomberg's exact-window 4.13 mb/d shows that vendor definitions differ enough to move the answer by nearly 6 percentage points (KSE Russian Oil Tracker, 1 Jul. 2026; Bloomberg via Meduza, 30 Jun./1 Jul. 2026).

The final gap is attribution. The fallback treats every barrel short of 5.0 mb/d as if it were unavailable capacity, while the actual shortfall can reflect normal underuse, crude output, grade mix, tanker timing, sanctions logistics, buyer demand, weather, or non-Russian barrels through Russian ports. That is why I treat 11.6% as the modal resolution value but 13% as the final mean forecast.